Dynamic pricing, vehicle relocation and staff rebalancing for station-based one-way electric carsharing systems considering nonlinear charging profile

ABSTRACT

This study focuses on the imbalance of vehicles in one-way station-based electric carsharing systems (OSECSs). A simulation-based optimization framework is proposed to determine optimal OSECS operation strategies, including dynamic pricing, vehicle relocation, and staff rebalancing. In this framework, the simulation module considers the nonlinear charging profiles for electric vehicles (EVs), EV-dependent road congestion, and price-dependent elastic demands to evaluate the performance of operation strategies. The optimization module triggered by the simulation model provides the optimized configuration of operation strategies based on the genetic algorithm and simultaneous perturbation stochastic approximationTo eliminate the decision variable combinatorial explosion for the optimization problem, we apply the K-means clustering algorithm to cluster stations into small groups of cluster zones, reducing the dimensions of decision variables. A case study in Chengdu demonstrates the proposed simulation-optimization framework’s efficiency and further analyzes the impact of different operation strategies, nonlinear charging mode, and EV-dependent road congestion for the OSECSs.

KEYWORDS:

• Carsharing
• nonlinear charging profile
• genetic algorithm
• simultaneous perturbation stochastic approximation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [No.71901183] and the Applied Basic Research Programs of Science and Technology Department of Sichuan Province [No. 2021YJ0066].